AutomaticParkingSystemANPR

This is an affordable solution using image processing of number plates to detect, identify and monitor vehicles in different scenarios. This is a project made for eliminating the need for manual registers and multiple watchmen in a simple cost-efficient way :) It uses Yolo v3's object detection to detect number-plates and follows it by tesseract OCR to identify the characters from the recognized number-plates. The android app is made using java. We use firebase for the database storage and also as the link between python and android codes.

A user registers his/her number plate before reaching the parking destination. One on reaching presses the enter button( to be implemented on the app itself. temporary solution, made a PyQt5 app on python.) On exiting the user presses the exit button which automatically calculates the total parking time, cost, etc. Payment structure yet to be implemented.

This was done for Smart India hackathon 2020 as well as Looking beyond Syllabus 2020.

Demo youtube video: -

Automatic Number Plate Recognition system - Automated Parking

Steps to run: -

1. Download THIS and add this to BackendPy folder.
2. Firebase Setup: -
   • Create a new project in firebase.
   • Add app android
   • Follow the steps given there. ( package name for this app com.example.database)
   • Create a new database( test mode )
   • Download the json file and use it to build a database on firebase.(import json option). It should then look something like this: -
3. Android Stuff: -
   • Open the android project in android studio.
   • tools->firebase->realtime data base-> connect to your firebase.
   • Follow the steps given there.
   • After everything, this file(google-services.json) should have been added in the app folder. Like This: -
4. Install requirements.txt for the backend python part.
5. Install the app on phone/emulator.
6. Tesseract Ocr stuff: -
   • Download and install this anywhere on your computer
   • Go to line number 88 in video_final.py
   • copy path till your tesseract.exe and paste it there(Line 88). Path should look like this D:\\softwares\\Tesseract-ocr\\tesseract.exe
7. Go to line number 84 and 85 and give a path to store temporary frames of the video being processed.
8. Open video_final.py and on line number 235(with ui) or on 246(without ui) give path to a video or keep path = 0 to activate webcam( use a printed number plate ) Sample video
9. Now inside the app, In the Advance(auth) button, select the floating button and add your number plate there.
10. To run without the PyQt5 ui directly run video_final.py and enter 1 for entry, then 0 for exit.
11. To run with ui, Run the design.py code.

Team X Æ A-4

Jay Jhaveri: https://github.com/JayJhaveri1906

Prem Chhabria: https://github.com/premchhabria

Abhay Gupta: https://github.com/abhay8463

Sahil Lotya: https://github.com/sahillotya

Rahul Koli: https://github.com/rahul2429

Prasad Govekar: https://github.com/govekarmohit